Recurrent circuits linking the cortex and basal ganglia in the mammalian brain mediate various types of procedural learning, such as acquisition of behavioral skills resulting in a habitual sequence of movements. Parallel cortico-basal ganglia circuits control distinct but related functions during acquisition of stereotyped motor habits: (1) evaluating specific behavioral actions in relation to a goal and (2) guiding motor performance to select appropriate actions. A fundamental question with regard to mechanisms of procedural learning is: how do neural circuits evaluate behavioral outcomes and use those evaluations in order to achieve the correct behavioral goal? Vocal learning in songbirds represents a type of procedural learning in which self-produced vocalizations are evaluated via feedback and progressively refined to achieve an imitation of memorized vocal sounds. This process requires the brain to compare feedback of current vocal behavior to the goal - the memory of vocal sounds. We have discovered that parallel CORE and SHELL cortico-basal ganglia pathways control goal-oriented versus motor-related aspects of vocal learning in songbird brain, thus providing a powerful model to study neural mechanisms of procedural learning that lead to stereotyped motor habits. We will investigate mechanisms by which parallel CORE and SHELL pathways support early stages of vocal learning as a model for how cortico-basal ganglia circuits mediate action-outcome evaluations. The proposed experiments will test the idea that SHELL circuitry is a primary site of comparison of behavioral feedback to the neural memory of vocal sounds, and that the results of this goal-oriented learning are necessary for accurate imitation of vocal sounds. We will record neural activity of single neurons in SHELL circuits in awake behaving songbirds as they are actively engaged in learning in order to test how evaluative signals are generated depending on whether behavioral feedback matches the goal memory of vocal sounds. We will also alter normal neural activity in SHELL circuits in juvenile singing birds to test whether disruption of evaluative signals impairs learning of specific vocal sounds. These studies will elucidate mechanisms of goal-oriented evaluation and how they influence initial stages of procedural learning. Investigations of procedural learning and its neural substrate are essential for understanding a variety of disorders involving stereotypic behaviors, including Parkinson's disease, Huntington's disease, obsessive-compulsive disorders, Tourette syndrome, autism spectrum disorders, and Rett syndrome.